1
$\begingroup$

Consider a Riemannian manifold $\mathcal{M}$ modeled on a (possibly infinitely dimensional) Hilbert space. Suppose that $\{p_i\} \subset \mathcal{M}$ and $p_i \to p$ .

We say that a sequence of covectors $v^*_i \in T^*_{p_i}\mathcal{M}$ is weakly converges to $v \in T_p^*\mathcal{M} $ if for every smooth function $f$ on $\mathcal{M}$ we have $$\langle v^*_i, df(p_i) \rangle \to \langle v^*, df(p) \rangle. $$

Question: Is it true that if $\{v_i^*\}$ is bounded then it has a weakly convergent subsequence?

$\endgroup$
1
  • $\begingroup$ When I corrected the title, I missed the obvious grammatical error in the very first word! [hangs head in shame] $\endgroup$
    – David Roberts
    Aug 6, 2017 at 11:24

1 Answer 1

4
$\begingroup$

Sorry, I did not read it carefully enough. Here is the revised version.

The answer is yes, if it is a strong Riemannian manifold modelled on Hilbert space (strong: generated the topology on each tangent space).

Namely, since $p_i \to p$ we may assume without loss that all $p_i$ are in one chart, so we are in a Hilbert space $H$, and the Riemannian metric is bounded by a contant times the standard one on an open subset containing all $p_i$.

By assumption, the $v_i^*$ are bounded, thus contained in a closed ball which is compact in the weak topology, thus there is a weakly convergent subsequence. Note that for each $(p,\alpha)\in T^*_p H$ there exists a smooth functions $f$ such that $df(p) = \alpha$. So we have the the weak topology to deal with.

$\endgroup$
0

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.